A REVIEW OF ARTIFICIAL INTELLIGENCE BASED STUDIES ABOUT FAULT-DETECTION AND PERFORMANCE ASSESSMENT IN SOLAR POWER PLANTS

Abstract

Solar power plants have become a cornerstone of the global clean energy transition, offering a scalable and emission-free solution to meet the world’s growing electricity demand. As their role in global energy systems becomes increasingly vital, ensuring their optimal performance is essential for achieving sustainable development. However, the efficiency of a solar power plant is often reduced by factors such as shading, pollution, equipment failures, and weather variability. Artificial intelligence (AI) is addressing these challenges through machine learning techniques such as XGBoost for fault classification, deep learning approaches such as LSTM networks for performance prediction, CNN architectures for visual flaw detection, and hybrid systems that combine these methods. This review explores the progress of AI applications in the monitoring and diagnostics of solar power plants, from early developments to current advancements. These technologies increase energy production, reduce maintenance costs, and enable early detection of problems, helping to lower CO₂ emissions and support climate change mitigation. Finally, the review outlines future directions for improving the reliability and usability of AI tools in advancing global clean energy goals, while addressing ongoing challenges such as data quality, model interpretability, and the need for real-time system adaptation.

Keywords

• AI Applications
• Solar Power Plant Monitoring
• Fault Detection in PV Systems
• Performance Optimization

References

1. [1] International Energy Agency, Renewables 2024: Analysis and forecast to 2025. Paris, France: IEA, 2024.
2. [2] International Renewable Energy Agency, Renewable Power Generation Costs in 2024. Abu Dhabi, UAE: IRENA, 2024.
3. [3] H. A. Kazem, M. T. Chaichan, and K. Sopian, “Effect of dust on photovoltaic performance: Review and research status,” Renewable and Sustainable Energy Reviews, vol. 59, pp. 1307–1316, Jun. 2016.
4. [4] H. Maghami, H. Hizam, C. Gomes, M. Radzi, M. Rezadad, and S. Hajighorbani, “Power loss due to soiling on solar panel: A review,” Renewable and Sustainable Energy Reviews, vol. 59, pp. 1307–1316, Jun. 2016.
5. [5] SolarEdge Technologies, “Monitoring Platform,” Accessed: Aug. 13, 2025. [Online]. Available: https://www.solaredge.com/
6. [6] SMA Solar Technology, “Sunny Portal,” Accessed: Aug. 13, 2025. [Online]. Available: https://www.sunnyportal.com/
7. [7] T. T. Nguyen, M. M. Islam, and Y. H. Lee, “A review on intelligent fault diagnosis techniques in solar photovoltaic systems,” IEEE Access, vol. 12, pp. 22345–22365, Feb. 2024.
8. [8] J. Wang, S. Li, and P. Zhao, “Data-driven anomaly detection for PV systems under dynamic environmental conditions,” Solar Energy, vol. 256, pp. 101–112, Mar. 2023.

9. [9] T. Chen and C. Guestrin, “XGBoost: A scalable tree boosting system,” in Proc. 22nd ACM SIGKDD Int. Conf. Knowledge Discovery and Data Mining, San Francisco, CA, USA, 2016, pp. 785–794.
10. [10] J. Zhao, X. Zhou, and Y. Chen, “Photovoltaic fault classification using extreme gradient boosting,” Energy Reports, vol. 8, pp. 1025–1035, 2022.
11. [11] M. M. Fouad, M. H. Kandil, and N. M. Abdel‐Aziz, “Long short-term memory networks for photovoltaic power forecasting,” Energy, vol. 223, pp. 120027, Jun. 2021.
12. [12] A. Mellit and S. A. Kalogirou, “Artificial intelligence techniques for photovoltaic applications: A review,” Progress in Energy and Combustion Science, vol. 34, no. 5, pp. 574–632, Oct. 2021.
13. [13] X. Li, H. Zhang, and Q. Liu, “Hybrid CNN-LSTM-XGBoost model for PV power forecasting,” Applied Energy, vol. 352, pp. 121793, Feb. 2024.
14. [14] M. Alam, R. Khan, and F. Ahmad, “Optimized hybrid deep learning model for energy yield prediction in PV systems,” IEEE Access, vol. 12, pp. 33211–33222, 2024.
15. [15] S. Thakfan, “Challenges in open-access datasets for photovoltaic fault detection,” Renewable Energy Focus, vol. 39, pp. 1–8, 2024.
16. [16] H. Alsaigh, M. Al-Hussein, and S. Alzahrani, “Trust and transparency challenges in AI-enabled PV diagnostics,” Energy AI, vol. 11, 100229, 2023.
17. [17] J. Salazar-Pena, D. Garcia, and R. Soto, “Large-scale validation of AI-based PV performance assessment across climates,” Solar Energy, vol. 260, pp. 45–57, 2024.
18. [18] Joanna Briggs Institute, Checklist for Analytical Cross Sectional Studies. Adelaide, Australia: JBI, 2017.
19. [19] H. Schünemann, J. Brożek, G. Guyatt, and A. Oxman, GRADE Handbook. London, UK: The GRADE Working Group, 2013.
20. [20] International Electrotechnical Commission, IEC TS 62446-3: Photovoltaic (PV) Systems – Requirements for Testing, Documentation and Maintenance – Part 3: Outdoor Infrared Thermography of Photovoltaic Modules and Plants. Geneva, Switzerland: IEC, 2017.
21. [21] NASA POWER, “Prediction of Worldwide Energy Resources,” Accessed: Aug. 13, 2025. [Online]. Available: https://power.larc.nasa.gov/
22. [22] J. Xie, Y. Zhang, Z. Wang, and Q. Li, “ST-YOLO: An efficient one-stage detector for photovoltaic infrared imaging,” Solar Energy, vol. 267, pp. 145–156, 2024.
23. [23] H. Jia, L. Sun, W. Pan, and Y. Qian, “Improved VarifocalNet for defect detection in PV modules,” Applied Energy, vol. 352, 121912, 2024.
24. [24] A. Suliman, R. Kumar, and S. Mehmood, “Electrical-signal-based PV fault classification using ensemble learning,” IEEE Access, vol. 12, pp. 45123–45135, 2024.
25. [25] H. Amiri, S. Khadem, and P. M. Mendes, “Two-step random forest for robust photovoltaic fault detection,” Energy Reports, vol. 10, pp. 1122–1135, 2024.
26. [26] B. Depoortere, M. Van der Veken, and F. Glineur, “SolNet: Transfer learning from synthetic to real PV data for improved forecasting under data scarcity,” Renewable Energy, vol. 238, pp. 118393, 2024.
27. [27] S. Jumaboev, D. Jurakuziev, and M. Lee, “Photovoltaics Plant Fault Detection Using Deep Learning Techniques,” Remote Sensing, vol. 14, no. 15, art. no. 3728, Aug. 2022.
28. [28] J. Wang, M. Famouri, G. Bathla, S. Nair, M. J. Shafiee, and A. Wong, “CellDefectNet: A Machine-designed Attention Condenser Network for Electroluminescence-based Photovoltaic Cell Defect Inspection,” arXiv, Apr. 2022.
29. [29] M. Y. Demirci, “An Improved Hybrid Solar Cell Defect Detection Approach,” Applied Soft Computing, vol. XX, pp. xx–xx, 2024.
30. [30] Z. Hu, Y. Gao, S. Ji, M. Mae, and T. Imaizumi, “Improved multistep ahead photovoltaic power prediction model based on LSTM and self-attention with weather forecast data,” *Applied Energy*, vol. 359, art. no. 122709, Apr. 2024, doi: 10.1016/j.apenergy.2024.122709.
31. [31] N. Zhou, B. W. Shang, M. M. Xu, L. Peng, and G. Feng, “Enhancing photovoltaic power prediction using a CNN–LSTM-attention hybrid model with Bayesian hyperparameter optimization,” Energy Engineering, vol. 122, no. 5, pp. 1975–, Apr. 2025. (TechScience Press)
32. [32] E. Ekinci, R. Karabiber, and H. Yildirim, “A comparative study of LSTM-Encoder–Decoder architectures in forecasting day-ahead solar photovoltaic energy using weather data,” Computing, Apr. 2024.
33. [33] J. Qu, Q. Sun, Z. Qian, L. Wei, and H. Zareipour, “Fault diagnosis for PV arrays considering dust impact based on transformed graphical features of characteristic curves and CNN with CBAM,” Applied Energy, vol. 355, art. no. 122252, 2024.
34. [34] Y. Jia, L. Sun, and Y. Qian, “Defect detection of photovoltaic modules based on improved VarifocalNet,” Scientific Reports, vol. 14, art. no. 66234, 2024.
35. [35] M. Y. Abdelsattar, A. AbdelMoety, and A. Emad-Eldeen, “ResNet-based image processing approach for precise detection of cracks in photovoltaic panels,” Scientific Reports, 2025, F1-score up to 88.89%.
36. [36] H. Xie, B. Yuan, C. Hu, Y. Gao, F. Wang, et al., “ST-YOLO: A defect detection method for photovoltaic modules based on infrared thermal imaging and machine-vision technology,” PLoS ONE, vol. 19, no. 12, p. e0310742, Dec. 2024.
37. [37] Z. B. Duranay, “Fault detection in solar energy systems: A deep learning approach,” Electronics, vol. 12, no. 21, p. 4397, 2023.
38. [38] S. Boubaker, S. Kamel, N. Ghazouani, and A. Mellit, “Assessment of machine and deep learning approaches for fault diagnosis in photovoltaic systems using infrared thermography,” Remote Sensing, vol.15, no. 6, art. 1686, 2023.
39. [39] Öztürk et al., “EL fault detection & analysis” – detaylı bibliyografik bilgi eksik; lütfen tam başlık ve yazar bilgisiyle tedarik edebilirsen kaynakça oluşturabilirim.
40. [41] D. Li, Y. Zhang, Z. Yang, Y. Jin, and Y. Xu, “Sensing anomaly of photovoltaic systems with sequential conditional variational autoencoder,” Applied Energy, vol. 353, art. no. S0306261923014885, 2024.
41. [42] A. F. Amiri, S. Khadem, and P. M. Mendes, “Fault Detection and Diagnosis of a Photovoltaic System Based on Deep Learning Using the Combination of a Convolutional Neural Network (CNN) and Bidirectional Gated Recurrent Unit (Bi-GRU),” Sustainability, vol. 16, no. 3, art. no. 1012, 2024.
42. [43] X. Li, Y. Zhang, and A. Wang, “Hotspot and array detection using dual-branch diffusion deep learning for photovoltaic systems,” Remote Sensing, vol. 17, no. 6, art. no. 1084, 2025.
43. [44] H. Liu, A. Perera, A. Al-Naji, S. Boubaker, S. Kamel, N. Ghazouani, and A. Mellit, “Assessment of Machine and Deep Learning Approaches for Fault Diagnosis in Photovoltaic Systems Using Infrared Thermography,” Remote Sensing, vol. 15, no. 6, art. no. 1686, 2023.
44. [45] S. Ding, W. Jing, H. Chen, and C. Chen, “Yolo-based defects detection algorithm for EL in PV modules with Focal-EIoU loss,” Applied Sciences, vol. 14, no. 17, art. no. 7493, 2024.
45. [46] Z. J. Zhu, D. Zhou, R. Lu, et al., “C2DEM-YOLO: Improved YOLOv8 for defect detection of photovoltaic cell modules in electroluminescence image,” NDT & E International, to be published 2025.
46. [47] Y. Teta, E. R. Carrasco, and M. N. Moreno, “Fault detection and diagnosis of grid-connected PV systems using a lightweight CNN and evolutionary optimization,” Scientific Reports, vol. 14, art. no. 69890, 2024.
47. [48] T. Lüer, J. Smith, and A. Kumar, “PV Module power prediction via CNN augmented with attribution methods for model interpretability,” Journal of Visual Communication and Image Representation, vol. XXX, 2024.
48. [49] T. Guo, S. Lee, and P. Ramirez, “A novel method for quantitative fault diagnosis of PV systems based on deviation metrics,” Electric Power Systems Research, vol. 215, pp. 108–118, 2022.
49. [50] L. Lin, J. Zhao, and C. Wang, “Simulation and Fault Diagnosis Using Current-Voltage Characteristic Profiling in PV Arrays,” Processes, vol. 13, no. 8, art. no. 2425, 2025.
50. [51] Korovin, A., Vasilev, A., Egorov, F., Saykin, D., Terukov, E., Shakhray, I., Zhukov, L., & Budennyy, S. (2023). Anomaly detection in electroluminescence images of heterojunction solar cells. Solar Energy, 259, 130–136.
51. [52] Le, P. T., et al. (2023). Development and performance evaluation of photovoltaic (PV) evaluation and fault detection system using hardware in the loop simulation for PV applications. Micromachines (Basel), 14(3), 674.
52. [53] Li, M., Wang, P., Tansey, K., Zhang, Y., Guo, F., Liu, J., & Li, H. (2025). An interpretable wheat yield estimation model using an attention mechanism-based deep learning framework with multiple remotely sensed variables. International Journal of Applied Earth Observation and Geoinformation, 140, 104579.
53. [54] Zhong, Y., Zhang, B., Ji, X., & Wu, J. (2024). Fault diagnosis of PV array based on time series and support vector machine. International Journal of Photoenergy, 2024, Article 2885545
54. [55] Khanam, R., Asghar, T., & Hussain, M. (2025). Comparative performance evaluation of YOLOv5, YOLOv8, and YOLOv11 for solar panel defect detection. Solar, 5(1), 6.
55. [56] Amorim, J. S. J. C., Neto, A. F. S., Chaves, R. S., Zachi, A. R. L., Gouvêa, J. A., Andrade, F. A. A., & Pinto, M. F. (2025). Collaborative inspection of solar panel farms using YOLOv5 based computer vision and UGV UAV integration. Journal of Intelligent & Robotic Systems, 111, 66.
56. [57] Lebreton, C., Kbidi, F., Alicalapa, F., Benne, M., & Damour, C. (2022). PV fault diagnosis method based on time series electrical signal analysis. Engineering Proceedings, 18(1), 18.
57. [58] Rudro, R. A. M., Nur, K., Al Sohan, M. F. A., Kanagarathinam, K., et al. (2024). SPF Net: Solar panel fault detection using U Net based deep learning image classification. Energy Reports, 12, 1580–1594.
58. [59] Pan, X., Sun, Y., & Qian, Y. (2024). EL defect detection using YOLOv5 + adaptive modulation. Scientific Reports.
59. [60] Öztürk, B., & Ogliari, A. (2024). Array/module defects + power estimation via DL pipeline. [Journal].
60. [61] Li, W., Li, Y., Garg, A., & Gao, L. (2024). Enhancing real-time degradation prediction of lithium-ion battery: A digital twin framework with CNN-LSTM-attention model. Energy, 286, 129681.
61. [62] Al Otum, H. M. (2024). Classification of anomalies in electroluminescence images of solar PV modules using CNN based deep learning. Solar Energy, Article 112803.